Kalahari San (G//anakhwe and G/wikhwe) in the Sedentary Community at !Koi!kom
著者(英) Kazuyoshi Sugawara
journal or
publication title
Senri Ethnological Studies
volume 30
page range 91‑116
year 1991‑03‑29
URL http://doi.org/10.15021/00003135
The Economics of Social Life among the Central Kalahari San (G/7anakhwe and G/wikhwe) in the Sedentary Community at !Koi!komi)
"KAZUYOSHI SUGAWARA
KIyoto Uhiver:sity
Hunter‑gatherer societies all over the World are currently undergoing prof6und transformations both in terms of economy and social organisation. The San of the harsh and dry Kalahari Desert, are no exception. In this paper I shall try to elucidate some essential features of the economic and social transformations of the life Of the GVanakhwe and G/wikhwe San in the sedentary community at !Koi!kom in the Central Kalahari Game Reserve, Botswana.
Since 1979, this area has undergone rapid change as a result of the Remote Area Development Programme (RADP) pushed forward by the government which has encouraged people to settle around the !Koi!kom borehole. Since 1981 under 'the guidance of the RADP, both the San and the Kgalagadi people living in other areas outside of tKade have begun to migrate to !Koi!kom. The people have settled into about 20 camps around the borehole and have become dependent on maize flour supplied by the government. Gathering of wild plants has declined, and the traditional bow‑and‑arrow hunting has been replaced by group hunting on horseback. This recent process of change in the life and society of the Central Kal,ahari San has been described in some detail by Tanaka {1987]2).
I will begin･by briefly examining group membership before turning to a consideration of the composition and derivation of people's belongings, including goats. I will then turn to q consideration of some aspects of social relations and cultural values in the sedentary situation. Drawing on conversational analysis3) I highlight some points of tension between the difiierent language speakers now thrown together and the way in which the perpetuation of the formalised extra‑
marital relationship underlines the continuing significance of economic mutualism.
1) The main part of this paper was originally presented at the Fifth International Conference on Hunting and Gathering Societies, Museums and Art Galleries of the Northern Territory, Darwin, 29th August‑2nd September 1988.
2) Cultural‑ecological studies were carried out there by George Silberbauer between 1958 and 1966 ISiLBER.BAUER 1972, 1981] with the G/wi San living in the gEKade area located in the mid‑western part of the reserve. In 1966, just after Silberbauer left, Tanaka Jiro started his research. Tanaka has studied the demography, subsistence ecology, group structure, nutrition and hygiene of the Glana and G/wi San within this area over a 40 month period [TANAKA 1976, 1980]./
91
I conclude by discussing the extent to which so called egalitarian values have been retained.
GROUP MEMBERSHIP
A population census in 1982 revealed that there was no significant difference in number among difiierent kinds of primary relative linkages, by which households were connected with one another within the same camp (Table 1). This result suggests, that the camps in the sedentary community are bilaterally organised, as was the case in their previous nomadic life [LEE 1979; TANAKA 1980]. Although camp membership has become far stabler after sedentarisation, occasional membership changes were observed." The process of fission and fusion of "family clusters" [TANAKA 1980] has been most closely observed among a group of three neighbOuring camps totalling 66 G/wi‑speaking people, including children, in the dry season of 1982. By 1984, this group had split into three distinct camps which were designated as P, S, and Km [SuGAwARA 1988] but by 1987, a tendency toward coresidence had become frequent again. For example Camp S and a large part of Camp P (Pi) are now located adjacent to each other (Figure 1). The total number of residents of the group exceeded more than 80 in 1987. This rapid population growth is mainly due to a rise in the birth rate and a conspicuous drop in infant mortality, the latter obviously caused by 'the excellent medical treatment offered;by a clinic established in 1983.
XBeginning in 1982, Tanaka has drganised four successive research projects on the effects of this rapid transformation on the life and society of the San [OsAKi 1984;
KiTAMuRA 1990; IKEyA 1989, 1991]. Participating in thiS project, during the first research period (August 1982‑February 1983) I started'a ̀socio‑ethological' study on the non‑verbal aspects of face‑to‑face interaction, proxemic behaviour and physical contact between ,individuals [SuqAwARA 1984]. Gradually my interest has shifted from the nonverbal or 'ethological aspects to the verbal or socio‑cultural aspects of face‑to‑face interaction; During the second re'search period (August 1984‑January 1985) I studied the patterns of visiting between camps, the interactions between residents and visitors and jn particular the gTeeting interactions {SuGAwARA 1988]. During the third period (July‑
November t987) I listed the property of each household in the camp where I was working, '‑and collected samPles of everyday cOnversation. In the most recent paper, I attempted to integrate observations accumulated during all three periods into an argument on the interactional modes ofthe ̀body' in co‑presence [SuGAwARA 1990]. Finally, during the last period (July‑November 1989) my analysis was concentrated on the conversational organisation and contents of discourse. The following description is primarily based on the data collected during the third research period,
3) Close examination of conversational interaction per se is beyond the scope of this chapter. Thus ･I cannot help omitting'many details, and summarising complicated materials into compact and inteligible narratives. Considerations on the interactional organisation or rhetorical features of discourse have been given elsewhere [SuGAwARA 1991, n.d.].
Table 1. Numbers of primary relative linkages among eqch camp.
adult residents of
Kinds of primary linkage
Dialect group GIana‑
prevalent G/wi‑
prevalent Mixed Total
Parents‑Son ' Parents‑Daughter Father‑Son Father‑Daughter Mother‑Son Mother‑Daughter Brothe.r‑Brother Brother‑Sister Sister‑Sister
2 6 1 3 11 10 5 11
8.
3 3 1
o 5 5 6 4
11
1 1
2 1 2 3 2 3 4
6 10 4 4 18 18 13 18 23
t
Kinship relations among residents of 13 camps were ascertained ip AugustlSeptember 1982. Sibling relations include those between half‑siblings by either father or mother. Statistically, no significant difference was found among Glana‑prevalent, G/wi prevalent, and mixed camps.
In 1987 intensive observations were focussed on three neighbouring camps, here designated as Pi, P2, and S. Fifteen ̀households' were distinguished within this group. A household is defined as a set of persons sleeping in the･same ̀hut' '
ttt .
'
1982 ABC DE KLM N FGH I J
xN NN
x
1983 ABC DE KLM ‑FGH XY I J
1984
1987
ABCO*
Km
/ tt
DE FG PQ*
P
KLMH I R
‑
x
‑ ‑‑'
J
tf l
ttll tt '
ABCO D
P,
EFG
Figure
PQ ST
P,
among a
xN NK
Ts
KLMHI
s group
u*v* JWZ
1. Fission and fusion of families of G/wi‑speaking people since
1982 to 1987.
Ovals indicate the camps and capital letters indicate the families composing each camp. Letters with asterisk indicate the new families formed by marriage.
14 2
4 1 3
12
5
6 15
8
10
11
Figure 2. Genealogy of adult members of the group in 1987.
14
13 7 9
Arabic figures indicate the households most of which correspond to the families
in Figure 1: 1=E; 2=K, 3=D, 4 :S, 5=F, 6=I, 7=L, 8= G, 9=M, 10=P, 11=V, 12=Q, 13=U, 14=H, 15=T. Nos. 1‑13 are in accordance with the relative age order among the husbands.
(ng!u:) at night and usUally sharing the same food. In San society, a household usually corresponds to the ̀nuclear family', which consists of a married couple and their immature offspring4).
Of the 15 households within the subject group, 1 1 consisted of married couples and'their offspring, while two consisted of young couples without offspring. Two more households consisted of ̀second' wives and their children, living in different huts from their husbands. One of them had left her husband in another camp for a long period, living with her married sisters in Camp Pi:
Figure 2 shows the simplified genealogy of adult members in the group who are or have been married. It should be noted that most households are connected to each other by primary relative linkages‑ (ties between parent and offspring or between siblings) and most members are from the G/wi dialect group. Only the wife of household No. 7 and the husband of No. 10 are G77ana‑speaking. The household numbers, with the exception of Nos. 14‑15, are listed in accordance with the relative age order among the husbands.
4) The major deviations from the equivalence between a household and a nuclear family:
(1) The position of adolescents: adolescents cannot sleep within the same hut as their parents. Adolescent males usually sleep outdoors, or behind a simple fence, near an open fire while adolescent females usually build their own hut, or gather in the hut of ̀widowed' or ̀separated' wives (see below). In both cases, they usually eat with their parents. Howev'er, there are many options for the adolescents as to where and with whom they can eat.
(2) The single parent family: single parent families result when one spouse has died or been absent for a long period, or when the couple have divorced.
(3) Polygyny: if a man is married to two or' more women, the latter build their own respective huts. The husband might sleep and eat with his wives in rotation, but only one wife lives with him at a time in the same household.
ECONOMIC ASPECTS OF THE SEDENTARY LIFE
It has been pointed out that San material culture was very simple: the kind and amount of property was limited to the minimum level by the necessity of frequent and distant movement [LEE 1979; TANAKA 1980]. As is expected, since
sedentarisation, not only belongings but also livestock have rapidly increased. In what follows I shall analyse the property held by people in order to elucidate the conditions of economic life in the sedentary community. First, I shall analyse the
̀catalogue of belongings' in each household, which was confirmed by intensive interview of the owner and second, the ownership of goats raised in the camps.
Composition and derivation of belongings ･
I asked the husband or wife of each household hQw he or she obtained each item they own (Photo. 1). A total of 884 items were recorded, of which 3.3 percent (29/884) were folk‑crafts to be sold to Ghantsi Craft, a corporation manqged by several European staff. Excluding these folk‑crafts, the kinds and derivation of 855 items of daily use were ascertained. The belongings are classified into seven categories: clothing (including shoes and sandals), bed clothes, containers (bags, bottles, cans, etc.), tools (hunting and gathering gear or handworking tools), utensils, hides (untanned or in the process of tanning), and other. The last category covers a wide range of goods, e.g., ornaments, pipes, ･walking sticks, odds and ends, and medicine.
The ways in which these possessions were obtained have been classified into nine categories: 1) g(ft: the owner (designated as Ego below) was presented with the item by other persbns. This category includes ̀hand me downs' (!kaoshina),
Photo. 1. Taking an inventory ofbelongings. An adolescent informant (left) is showing his father's belongings such as mortar, long cooking stick (fo'r stirring meat), 'axe, quiver, carrying net, cup, pot, water container, dishes, etc.
usually clothes to Ego. 2) barter: the owner ̀exchanged' (duera) some goods in return for the goods received. This category includes the cases of ̀equivalent exchange' (tsentsa) where the same kinds of goods are exchanged only for the sake of preference. 3) purchase: Ego ̀buys' (Iden) the good with cash.' 4) reward:
goods received as a reward for his or her ̀labour' (tsei 5) found: items ̀found' ( 7E kxderi or ko':) while walking around, which had been thrown away by somebody else. 6) se(f:macle: the goods ̀made' (tsaon) by Ego. 7) hunted/gathered: goods Ego obtained (game or harvest) from hunting‑and‑gathering. Three methods of hunting, i.e., equestrian hunting, spear hunting with dogs, and snaring, were specified in the interview, but are not distinguished here. 8) rations: goods Ego received which were ̀rations' (haba) provided by the government, and 9) borrowed:
the goods Ego had ̀borrowed' (/o'be) from other' persons.
The most prevalent source of possessions is to receive it as a gij7 as opposed to receiving it by barter, or reciprocal exchange which was rare. Admittedly this method of synchronically enumerating the derivations of belonging is open to the danger of confusing cases of getting an item for nothing with that of "delayed reciprocity,'r where the return for a gift is to be paid sometime later, However, the San themselves are ･willing to declare that someone has ̀given' something ̀for nothing' (kx?amha ･ki'e), and'that anQther person has immediately asked for a return of another item. In the latter instance, the informants often say they have ̀bought' the good by ̀paying' something for it. Therefore it might be presumed that the gij7 partnerships represent some kind of positive social relationship, such as athnity, friendship, or an emotional attachment, while partnerships involved in both categories of barter and purchase represent more depersonalised or practical relationships. This presumption wi!1 be tested below, examining the range of partners involved in these kinds of transactions.
Several point's. emerge from a compatison of the ways in which the four categories o￡belongings were obtained (Figure 3). The most striking feature is that nearly 50 percent of the ･tbqlS h"av, e been made by the owners themselves. This fact suggests that thg subsistence cffor,ts of each household are made independently.
This supposition is suppd.rted by,the conspicuous infrequency of borrowed goods.
Tanaka [1980] pointed out ･that lending‑and‑borrowing of hunting ge'ar and other daily necessities were cgmmon during.'previous periods, and.'''argued that the willingness to le'nd and borrow' such itenis might be q sociological device reinforcing the principle. of reciprocity;'emphasising the interdependence between different households., The high. frequency of seij:made tools 'and the conspicuous infrequency of bOrrowed'6elongings seemsn to reflect the profound transformation of socio‑economic life from the previous system described by Tanaka.
Bed clotheS Standiin sharp contrast to tools, in that nearly 50 percent of the former were rationed out by the government. Bed clothes, especially blankets, are critical items which protect the aged, young, and sick from the severe cold in the dry winter. The blanket also seems to function as a kind of psychological shelter which imparts comfort and assurance to infants and children. It is one o'f the most
o 20 40 60 80 100%
Bed Clothes (114)
Clothing (308)
Utensils (68)
Tools <110)
Reward
Gift"
i':'i:'
:i' :･l:
':'i':1.:
‑‑‑‑‑‑
t‑‑‑ ‑‑‑‑‑‑ ‑+ ‑‑
‑‑‑‑‑‑‑‑‑‑‑‑‑
‑‑‑ ‑‑ ‑‑ ‑‑ ‑‑ ‑‑
‑‑ ‑‑ ‑‑ ‑‑ ‑‑ ‑‑
‑‑‑‑‑‑‑‑‑‑‑‑‑
‑‑‑‑‑‑‑
.
Barter"
i'""..$.gesu,.il'.'i"ii'ilgiitst/,/¥E#t/k:'i{"iis,,
l'S･i･･il/li/11:･,l/il･l/i･,,ij//i,iii･ig･･i/ii/r,:･:ii{･{:ii/iI'11/tl',killltr,'k'/Eilli,,ii.,/i･]t･/t･l･l･,i:‑/l･,:':S;.i,2･l.:liili･,E
‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑
iiiiiiIiiiiiiiiiiiiiiiiiii:iii"ii･iiiiii･iii:iiiiiiiiiiiiiiiii,iii :I:1:l:1･1:1:1:l:I:1:l:1:S‑sltt;M.,.g.9.g.I･1',･1',1IIlI11l':E1lI1II‑‑‑‑‑i‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑b‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑
Figure 3. The ways in which four categories of belongings were obtained.
Figures in the,parentheses indica'te the numbers of recorded items.
attractive and valuable goods, which people often buy for cash, even though a considerable number of them are supplied by the government. An examination of the distribution of the prices ofpurchased goods, reveals a distinctive clustering in the low price range: about 56 percent of purchases are included within the range of O.5‑5.0 pula (1 pula==ca. O.8 US$). However, the second most frequent price cluster lies within the range of 10‑15 pula. Most of the items included in this cluster are blankets. Especially, at the shpp near !Koi!kom borehole, blankets are sold at a fixed price of 12.5 pula. ･ This is a rather high price by San standards, but those who regularly obtain cash income do not hesitate to buy new blankets. A considerable number of bed clothes have been obtained as, g(frs and it was often the case that people received ̀hand me downs' (!kaoshina) of old blankets pierced or '
scorched with holes. ' '･
The people get about 77 percent of their clothing as gij7s or as ration in equal proportions. In the present sedentary community, clothes are regularly distributed at the clinic by the missionary nurse, so that the people seein to have enough clothes. Previously Western style clothes were highly yalued by the San but nowadays the scarcity of clothes has decreased, and their relative abundance enables the people to readily use them as gifts which serve to confirm positive･sQcrial relationships.
The ways in which utensils have been obtained are quite variable: they might be received as gij7s (29.4%), by purchase (26.5%), barter (19.1%), or seij:made (17.6%). The prices of utensils also cover a wide range: most were purchased at the price of less than 10 pula, but one conspicuous exception was observed, in which a big iron three legged pot was purchased at,the price of 60 pula. , This is the highest price for one item of all c.ases recorded. The high value put on the three‑legged pots emerges from a consideration.of barter. Five pots were bartered for other items; 1) one male goat, 2) two female goats, 3) one male and four female goats, 4) a piece of caracal hide and 10 duiker hides, and 5) 20 steenbok hides. Thus, it can be concluded that the three‑legged pot is a very scarce and necessary item of daily life, for which the people have to pay a considerable amount of goods or money. Other cases of barter consisted of the exchange of ̀valuable' goods, e.g., a spear for a goat, a saddle for a donkey, etc. It can safely be concluded that bartering is a principal means by which the San obtain a highly desired and relatively ̀expensive' (duru:) good.
There is a striking variation among the households in the ways in which goods are obtained. This variation is primarily due to the different subsistence options which closely correlate with the husband's age (Figure 4). Thirteen husbands were divided into three age grades; old, middle, and young. The ̀old' husbands differ from the other husbands, in that they seldom purchase good for cash and are reluctant to engage themselves 'in folk‑craft making, the most reliable way to earn money. In contrast, ̀young' husbands differ from the others in that they rarely get their goods by means of hunting‑and‑gathering. They engage themselves in construction labour and regularly earn a considerable amount of money. Even
o 20 40 60 80 100%
Reward Old
(223)
Middle
(421) ,
'YQung
(172)
Figure
Gift
Found
･g
'i' rv"'i
:‑:::.;s.･:'i';i･i･:::'･::･‑
‑ ‑‑‑‑‑‑‑‑‑‑‑‑‑‑
‑.‑‑‑‑‑‑‑‑‑‑‑t‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑
‑‑‑‑‑‑‑ ny t‑t‑‑t‑‑:':l:1:1Self‑MadeI11I .II
‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑t‑‑‑‑‑‑‑‑‑‑‑‑‑4‑‑‑‑‑‑‑‑‑‑‑‑l‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑‑
･ Rationed
:1:l,il::.,l･ll･l'iiElll,}..l.lllll,l･llll,{,il'll･ligj･i･illl
IS//rPurchase;I.s･l
l/l･li,i･l,il'I,l'/gl･i,i/i･l.lli,lff},ll･},ti,Il,l･lli.i･i,II:,
Hunted/
Gathered
vvvv
', ･. Borrowed
4･ Variation in the ways in which belongings･ were obtained by three age grades of husbands (see the notation in FigUre 3).
ileen'l.",l.
Barter
though they sporadically participate in the equestrian hunting, they are almost indifferent to snare hunting which is the most reliable means by which.the skins of small antelopes, i.e., steenbok (!gden) and duiker (ng!owa), can be obtained. The indifference of younger men to such traditional activities in the bush may be an omen of future,diMculty the San will face in passing on their unique culture and knowledge of the environment to the next generation.
The partners who were involved in such transactions as gij7‑giving, purchase, and barter, are classified into five categories from the viewpoint of each informant:
consanguines, primary and secondary aMnes, non‑kin of G/wi‑speaking group, non‑kin of G77ana‑speaking group, and Kgalagadi. Three kinds of dyads, i.e., male‑male, male‑female, and female‑female are distinguished. Barter and purchase rarely occurred between males and females or among females but was most frequent between males amounting to 71.6 percent (58/81) of the cases. A significant difference was noticed between male‑male and female‑female dyads in the distribution of kinship or social distance between the owner and those who have presented the gij7 (Table 2). That is, males more frequently received gijts from partners of the same sex belonging to other language groups (GVana and Kgalagadi) than females, while females more frequently received gij7s from their female consanguines (chi‑square test for two independent samples: z2==20.5416, p<O.OOI, df==4). The parallel difference between the sexes in the range of social relationships was exemplified both for the proxemic relationships within the camp.
[SuGAwARA 1984] and for the visiting relations 'between camps [SuGAwARA 1988]. It can be concluded that social and economic relationships among females tend to be confined to those among close kin, while male relationships tend to be extended to a wider range of people.
Comparing kinship or social distance between the male partners involved in gij7‑giving relationships with the distance inpurchase and barter, we can recognise a slight tendency for consanguines to be involved more frequently in gijtr‑giving than in barter andpurchase (Table 3). A reverse tendency is recognis,ed for the partners who are only distantly related (see the columns for ̀G/wi' and ̀GVana' in Table 3).
But this difference between both kinds of transaction is not statiStically significant (x2==4.7461, n.s., df=4). The factor which diminishes the difference between these
Table 2. Kinship and social distance between givers and receivers of gij7s.
Dyad Consanguines Affines G/wi G"ana Kgalagadi Total
Male‑
male Female‑
female Total
19
(22.1)
A
32
(45.7)
51
(32.7)
36
(41.9)
24(34.3)
60
(38.4)
8(9.3)
11
(15.7)
19
(12.2)
6(6.9)
1
(1.4)
7(4.5)
17
(19.8)
Y 2
(2.9)
19(12.2)
86(1OO.O)
70(1OO.O)
156
(100.0) Figures in parentheses indicate the percentage to the total number in each row.
> : p < O.02, >> : p < O.Ol (x2‑ test for two independent samples, df = 1)
Table 3. Kinship and social distance
.
transactlons.
between male partners involved in two kinds of
Transaction Consang.' Affines G/wi GIana Kgalagadi Total Gift‑
givin'g
Barter &
PuTchase Total
19
(22.1) =.
6
(10.3)
25
(17.4)
36
(41.8)
t
26(".8) 62
(43.1)
8(9.3)
7
(12.1)
15
(10.4)'
6(7.0)
8
(13.8)
14 (9.7),
17
(19.8)
11
(19.0)
28(19.4)
86
(100.0)
58
(1OO.O)
1"
(100.0) See the notation of Table 2. (>) : p < O.10
two kinds of transaction is that tbe categories of ̀aMnes' and ̀Kgalagadi' respectively amount to almost the same percentage in both transactions. The category of aMnes, in particular, desierves closer examination.
The people in the tKade area are connected with Qne another by complicated and extensive conjugal networks, so that from any individual's point of view those included in the category of prima'ry and secondary aMnes amount to a great number. In 1984 I classified most of the San people living at !Koi!kom according to the kinship. distance from the adult members of camp P (Figure 1). Of 82 men for whom kinship relations were confirmed, 61 percent (50/82) were included in the category of primary and secondary atfines from the viewpoint of at least one of the
male residents of Camp P [SuGAwARA 1988: 182]. In order to clarify the correlation between social distance and nature of economic transaction, the variation within this too large category of ̀athnes' was examined. For this .purpose, I divided the category of ̀aMnes' into two subcategories according to their residence (Table 4). The category of ̀co‑members' refers to a set of people who were living in the three adjacent camps Pi, P2, and S, in 1987 (Figure 1). Most of them were, of course, the･.subjects of this interview research on their belongings, though several unmarried adolescents are also included. We can recognise a sljght tendency for co‑members to be involved more frequently in gij7Tgiving than in barter andpurchase. But, again, this tendency is statistically not at all significant
(x2=1.9644,n.s.,df±1). ' '
Table 4. Residential relation between male partners who were involved in two kinds of transactions and included in the category of primary or secondary affines.
Co‑members Others Total
Gift‑
giving Barter &
Purchase Tbtal
23 (63.9)
12 (46.2)
35
13 (36.1)
14 (53.8)
27
36 26 62 Figures in parentheses indicate the percentage to the total number in each row･
Thus, the above assumption that barter and purchase represent a more practical and depersonalised relationship, than gW‑giving, cannot be supported so far as male‑male relationships are concerned. In other words, in the present sedentary community a wide‑ranging gift‑giving network is prevalent among males, especially between affines or between San and Kgalagadi. On the other hand, they seem to be oriented toward the profit‑seeking principle of expecting immediate return, even from their close kin. The ̀classical' type of gift exchange most frequently occurs in the explicit joking relationships, e.g., between an adolescent male and his mother's brother. A few cases of gift exchange were 'observed in persistent extra‑marital sexual relationships (za:kuaha) as well. More broadly, however, it can safely be concluded that consistent relationships of gift exchange with a specific partner, comparable to the hxaro relationship among the !Kung San
[WiEssNER 1982], are r'arely found among the Central Kalahari San.
Ownership, of goats5)
Since the late sixties there have been several ̀rich men' keeping many goats in the #Kade area, especially one GVana‑speaking man, born in Gyom. Currently he owns more than 300 goats, and exerts a great influence on the system of goat‑
raising in the community. Camp S is connected with this' ̀great goat owner' by aMnal relationship and of their herd of about 90 goats, one‑third belong to him. As is shown in this case, a large‑scale goat owner is willing to entrust the management of a part of his herd to other camps. Similarly, anyone who owns only a few goats also tends to entrust them to his or her close kin living in another camp who relatively speaking already have many more goats., As a result, a complex network of keeping‑and‑entrusting relationships for goats has'spread through the communlty.
An extended family including an old father (KK) with two wives apd his two married sons (SH and HX) constitute the core of Camp S (Figure 5). One of his sons (SH) is married to the daughter of the ̀great goat owner' (TS). The son of･the second, wife of KK (GR) also lives in this camp. KK and his sons own 40.9 percent"
(36/88) of all the goats in this camp, while GR owns only 6.8 percent (6/88).
Within the extended family ownership is not so clearly distinguished. When asked,
"Who owns this goat?," both the father and sons immediately answered, "This is mine." However, the five goats owned by the wife of SH are clearly distinguished from those of her husband and in‑laws, because she received them directly from her father. This separate ownership of property by the.wife from her'husband can be regarded as a symptom of subordination of the latter to his father‑in‑law. In fact, SH sometimes stated that he was very afraid of his father‑in‑law, and I have
5) The management･of goats was described in detail by Ikeya Kazunobu whose research focussed on a small camp of Glana‑speaking people during the period of 1987‑1988 [IKEyA 1991]. Here I do not intend to analyse the details of goat management by the San. Rather, I will point out some essential features of goat ownership.
o
as
hhA AA
10
20
30'
40
l･:lg.l
Ts
l".
I l l I l l I 1 l l
1
l I l l l
l
IL
$.E.i
I/l.:.:i.i/
:･2･}, ･
s
:
1
: : : : : : : :
1
: : : l : :
'
: :
"" ･:
P, : : : : : : : : : : : l : : : : : :
1 ."
M Adult ¥ tw Young ¥ Z Young g
‑ Castrated g
Others
Figure' 5. 0wnership Of the goats in Camp S.
Thg bars represent the number of goaSs owned by each individual in the uppermost section of each column. Hatched circles and triangles represent those who do not live in this camp. Letters at the bottom indicate the camps in which owners live.
observed him (SH) very flatteringly excuse himself for his bad manners after being mildly reproved for them by TS. Another 13.6 percent (12/88) of the goats in Camp S belong to seven non‑residents other than TS: six were owned by TS's half brother and another six owned respectively by six other men.
Camp Pi manages about 50 goats, most of which have been recently purchased by the residents who received cash income, mainly by selling folk crafts to Ghantsi Craft. Within this camp ownership of these goats is subdivided among 10 residents (3.9 head per person; range: 1‑10), including unmarried adolescent males. Only 13.3 percent (6/45) are entrusted to this camp by three non‑resident owners (Fjgure 6). The striking, feature concerning the recognition of goat ownership common to both camps, S and Pi, is that most people are quite indifferent as to which goat belongs to whom. In Camp S, KK, SH, HX, and the children who everyday help them with management have an accurate knowledge of ownership, but GR ' identifies only his own goats. In Camp Pi, three adult men, each of whom owns more than 6 goats, have a relatively accurate knowledge, but even they are sometimes uncertain as to the ownership of the goats of others. The colours and patterns of hair are varied enough that individual identification is possible, if one wants to do so. For most bf the San, goat‑raising is quite a new method of subsistence. It seems that they have not yet developed any sympathetic feeling for
the animals on which they have begun to become dependent. WhenIasked
o
5
10
P,
' ･e i"‑
:'.x ec:
: : : : : : : : : : : : :
1
Others
Figure 6. 0wnership of the goats in Camp Pi (see the notation in Figure 5).
whether the individual goats were named, or whether the life stages of the goats were distinguished by any named classes, TS, the great goat owner, laughed at my question, answering, "They are only goats. They have no ears. What do you want to do by naming them?" Thus, the San remain ̀hunter‑gatherers' at least in their attitude toward livestock6).
The most crucial question concerning the social arrangement of goat management is what are the economic and social functions, of entrusting one's own goats to other camps? On the face of it a caretaker does not receive any apparent benefit by keeping someone else's goats. Even if a female goat gives birth to many offspring, none of them are offered to the caretaker as a reward for his or her labour but remain the property of the goat owner; The only definite reward is their milk.
However the amount of milk produced is negligible especially during the dry season. On the other hand, the cost for keeping such goats is also negligible, because livestock management by the San is far from intensive. Every morning, , after separating the infants from their mothers, a few residents, sometimes only children, drive the herd away from･camp. Nobody follows the herd, as they spontaneously forage on their own around the community. While it is true ,that there is little reason to refuse the request to care for someone else's goats, it is also difficult to specify a definite economic purpose for the practice of caretaking goats.
Although further systematic research covering all the goat owners throughout the 6) In 1989, I have collected a conversation sample in which an immature goat, whether male or female, is referred to by a specific term hxo'ba:, which might have been imported very recently from the agro‑pastoralist vocabulary.
community is required, it is probable that caretaking goats might be one of the means by which the owner and the caretaker reathrm alliance or friendship between themselves.
SOCIAL RELATIONSHIPS AND CULTURAL VALUES
In this section, I shall deal with several topics extracted from the samples of everyday conversations, which are especially releVant to essential characteristics of social life in the sedentary community.
Relationships between ethnic or language gtoups
A large part of the populatiQn presently at !Koi!kom are immigrants from homelands other than #Kade. Some are derived from the Kgalagadi ethnic group.
Various symptoms of social tension and confiict between different ethnic or language groups can be recognised. The following is one example.
In the dry season of 1984, just before the research team arrived at gE Kade, the daughter of a Glana‑speaking man (NA) died suddenly, probably of an illness.
After the burial, a rumour began to.spread that her death was caused by a Kgalagadi sorcerer (RS), a rich man who had migrated from Metse‑a‑Manong about two years before. ･ Within a month, the rumour developed into a dangerous situatjon: The people said that the sorcerer intended to dig open the grave of his victim, and insisted that he should be arrested and be judged in court. The rumour gradually died down, and the matter seemed to rest there. However, about half a year later, the rumour about the same sorcerer, RS, was revived by NA following an extraordinary event. NA had been one of the best friends of Tanaka, director of our research team, and a member of the team, Kitamura, and his wife had lived in NA's carnp, building a nice hut as.a refuge from the intense daytime heat. One afternoon in later January 1985, just one or two weeks after we left =Kade, this hut was struck by lightening, and more than ten goats owned by NA, which were resting in and around the empty hut, died. NA was so horrified by this calamity, which was also ascribed to sorcery by RS, that he migrated to another camp.
On various occasions, the informants told me how RS carried out the sorcery.
It was said that on a dark night he stripped stark naked, stalked.up to the target camp, and threw some kind of medicine at the hut. It is noteworthy that this image of a sorcerer depicted by the San is very similar to the Bantu stereotype of the sorcerer which is quite Prevalent among the agriculturalist societies of Eastern or Central Afri'ca [NAGAsHiMA 1987]. The idea of sorcery itself might not be the invention of the Bantu but the prevailing idiom in which pepple talk about sorcery is now heavily infiuenced by the Kgalagadi and other Tswana tribes. I could not judge whether the San really believe in the magical poweir as the cause of the inscrutable calamity, or whether ,their explanation was merely a means by which they expressed distrust and antipathy toward the Kgalagadj. The important point is, however, that the San are tempted to put an emphasis on the difference between
themselves and the Kgalagadi by insisting that the latter have mysterious power or strange manners beyond comprehension.
Analysing the conversation samples, I was impressed with several cases in which the incomprehensible behaviours of not only Kgalagadi but also of Japanese and European people were vividly depicted. In one case, a man (NK) complained that his dog was "dying." When he passed by a Kgalagadi camp with his dog, the host of the camp set his big dog on NK's dog, which was covered all Qver with bites.
Moreover the host struck NK's dog with a club. Another man commented on this story, saying, "Why do those who have come from there kill other people's property? They are very Negro‑like people, aren't they? They are not right . We treat the dog in this way. We don't know it (i･.e., to beat). Even if a dog comes and steals from us, we go to the owner and tell him next morning. We never kill the dog."
The above discourse typically shows the way in which the San discriminate between themselves and the Kgalagadi. The "Negro‑like"7) people do not have the proper manners. They behave fiercely. They do not recognise how terrible it is to destroy another person's property...etc. Needless to say, the San themselves do not hesitate to beat their dogs. In fact, on another occasion I recorded a conversation in which seyeral men joyfully discussed effective ways of repulsing a barking dog. Thus, emphasis may not necessarily be put on the real difference between the San and other people but any trouble which one experiences in an encounter with others from another group can be interpreted as a good index of the ethnic character of the group itself.
The following example shows a more extreme case, in which an argument about the.ethnic character of another group is grounded.on an apparently far fetched rumour. An old woman told a story to a woman of middle age, about a Kgalagadi man using magical medicine. He supposedly cursed his younger brother and let the latter die. The middle aged woman then told another story. Long ago, the young sontof a Kgalagadi man named G/ui disappeared. The people looked for him, crying, "G/ui's son has been lost in the bush!" But G/ui himself was calmly sitting in a chair. Then he disclosed, "He isn't loSt. I haye sold him to the Government . I stamped the fingerprint for contract . " That boy was very cute and young. Now how tall is he? G/ui had many children. When they have grown up, they gave birth to many grandchildren. G/ui will sell them again...etc.
According to the San, this is the way the Kgalagadi get much money, and the reason why they are so rich. The above story reveals several means of essential characterisation of ̀we' and ̀they.'･ First, for the San, brotherhood is the most reliable tie between two individuals, but the Kgalagadi might kill even their own brother. Second, for the San the love towards one's own children constitutes the profound delight of life, while the Kgalagadi are cold enough to sell their own children. The third point might be more trivial, but it may be noteworthy as a 7) The San word iEtebe used for Bantu has slight negative overtones.
"body idiom" {GoFFMAN 1963], i.e., the Kgalagadi always sit on chairs, while San sit on the sand. Probably, we can add many other contrastive pairs to this list of
̀we'‑̀they' characteristics. The more valued the feature contrasted, the more incomprehensible and striking are the manners of other people ' which are characterised in terms of this feature. The San's discourse emphasising the difference between themselves and' other people rightly refiects the actual power relationship in which the San hold an inferior position. If they are involved in any trouble with the Kgalagadi, they are very eager to commit the matter to public judgement (shie:ka) in court (h/o:). However, this practice itself has been recently imported from the Tswana institution under supervision of the government. In an informal discussion within camp, the party who suffers any damage always insists that the other party is responsible for this damage, and that they should ̀pay' (duera) a penalty. But, most of the trial time is spent in confirming the actual course and details of the matter, and an effective consensuS hardly develops: a concrete penalty is rarely attained especially if, the damage is not material (Photo.
2). Finally, in spite of the bitter backbiting within camp, the San usuallY behave very temperately, or sometimes even fiatteringly, in actual face‑to‑face encounters with the Kgalagadi.
San men sometimes visit the camps of Kgalagadi, expecting a reward for doing temporary jobs such as processing hides or building fences around the fields [SuGAwARA 1988]; It is supposed that such a kind of ̀service' by the San to the local agropastoralist is not unique to the recent sedentary life, but has been occasionally practiced in the past (cf. Solway and Lee [1990]). The San are aware of a proper way to work for ̀Negros' (= tabe). This was brought out in a story about an old G/wi‑speaking woman living in another camp, who was criticised for
, mo11･EI5E. l:.,･ i
'.,･Illllll ;:,i
tw<:I"..k"msc
ss patw
Photo. 2. A scene of discussion in the court. A GIanakhwe man (right), standing face‑
to‑face with the appointed chief (left), is testifying that he saw a Kgalagadi man do violence to a G/wikhwe boy on the previous day.
her bad manners in offering a service. Recently she had moved to a camp of the appointed chief (a rich Glana‑speaking man) and shortly after had said to the chief and his wife, "Bring me your dirty things. I will wash them for you." A male commentator on this story remarked "How (stupid) that woman is! We never talk big in that way" and a female commented saying, "Negro people would offer something by themselves. It is not until they give us their things that we work for them." When analysing this story my informant further explained as follows:
traditionally, G/wi people used to visit the Negro with collected food. The host would eat the plant food, and, after eating, take out some goods from the house by himseijeso as to offer them to the G/wi visitor. Receiving them, s/he would repair, wash, sew, tan, those goods as was appropriate and return them to the host. The host would then pay some reward in token of his gratitude. Working in this fashion is referred to by the verb tsawa‑md which is a compound of ̀make' (tsaon) and ̀give' (md). This manner of working idealised by San themselves reveals essential features of the socio‑economic relationship between San and Kgalagadi.
The San can not enforce their demands on the Kgalagadi, and have to wait for the voluntary favour by the latter. At the same time, Kgalagadi are expected nQt to refuse visiting San, and to deal with them generously. Such traditional attitudes and expectations between the two ethnic groups may be maintained even under present conditions of close contact between them even though the relationship between the San and the Kgalagadi is open to antagonism and conflict. Both groups, however, are much more dependent on each other now, given their permanent co‑residence.
Another type of economic transaction between the San and Kgalagadi can be formed for an especial reason; i.e., ritual. On various occasions, the San are willing to ask a Kgalagadi ̀doctor' to carry oUt a ritual. The following is an example: it is believed that ̀breach birth' (kyaba:ma) causes the father to fall sick.
Following a breach' birth the father was separated from the mother and the newborn, and left to sleep in a separate hut. An old man, a classificatory ̀father' (FB) of the mother of the baby took the initiative in arranging for the healing ritual, as the father himself was unwilling to pay much for such a service. This led to a big argument with his elder brother and ̀father'‑in‑law. Finally the old man persuaded his ̀son'‑in‑law to go along with the ritual and succeeding in finding an old Kgalagadi man to conduct the ritual. The ritual was held and the family was healed so the father moved back to live with ･his wife and the newborn again in the same hut (Photo. 3). Thus, in spite of the San fear of the magical power exerted by the Kgalagadi, they are also attracted to it: such attraction is leading to more intense economic transactions between the two groups.
The San themselves are not homogeneous'. Ihave observed several cases in which some of the G"ana‑speaking people expressed a sense of superiority over the G/wi‑speaking people. Similarly, G/wikhwe sometimes make negative comments on the character of Gi;ianakhwe in their everyday conversation. The husband of household No. 10 is Glana‑speaking. His mother‑in‑law (household No. 5)
Photo.
lllllilll
Iiii ¥‑;: ": x ,. .,
ee
3. A Kgalagqdi ̀doctor' is eonducting a ritual which purifies the pollution caused by a breach birth. He is about to thrust his head and arms through a hole made on the back side of the hut to rub some medicine on the breast and belly of the newborn's mother sitting within the hut. It is assumed that this arrangement symbolically cortects the ̀wrong direction' in Which the baby was born.
complained about her 7years old grandson who would often tease his younger brother, saying, "He makes a whip and beats him (younger brother). This is the way, GVana child behaves. He doesn't distinguish his own younger brother (from other children)." This woman's criticism of her grandson (and GVana children in general) might be backed with ill feeling towards her GVana‑speaking son‑in‑law.
On another'occasion, this same woman quite plainly expressed her hostility toward the son‑in‑law named GB. She was talking about an argument which she recently had with a G/wi‑speaking man named KY living in a neighbouring camp. "That man with whom I talked, KY, doesn't have a heart like GB's. His heart is clean as if it were washed. My ̀younger brother,' the G/wi man, his heart is beautiful. He never says ̀ieshiq iesht'e' as Gi7anakhwe do." The interjection ieshie (̀what!?. ') is Gl;iana locution which approximately corresponds to G/wi usage of ihosi. Thus, by framing up the diflierence in everyday usage between two dialects, the speaker was effectively emphasising the displeasing character of,GVanakhwe: In this context she might have implied that, though KY a G/wikhwe, was to blame for his insincerity, he was not worse than GB, Gi7anakhwe.
G/wi‑speaking people also like to criticise Glana‑speaking people for their disabilities. A young man complained that, though his GVana‑speaking friends had patrolled to inspect his snares instead of himself, they had not set snares again after catching two steenboks. In response to this complaint, another young man said, "GIanakhwe don't know how to set a snare. Where do they come from?"
The locution of ̀where do you come from?? (tse ma:da hoa:ha?) is often used to
